584_test2

This model is a fine-tuned version of bert-base-uncased on the None dataset. It achieves the following results on the evaluation set:

• Loss: 1.6299
• Accuracy: 0.7625

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 5e-05
• train_batch_size: 32
• eval_batch_size: 64
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	121	1.2163	0.4771
No log	2.0	242	0.8698	0.6708
No log	3.0	363	0.8224	0.6792
No log	4.0	484	0.7631	0.7208
0.9113	5.0	605	0.8396	0.7146
0.9113	6.0	726	0.9844	0.7
0.9113	7.0	847	1.1391	0.7063
0.9113	8.0	968	1.0962	0.725
0.1905	9.0	1089	1.2468	0.7396
0.1905	10.0	1210	1.4225	0.7292
0.1905	11.0	1331	1.4732	0.7396
0.1905	12.0	1452	1.5534	0.7438
0.058	13.0	1573	1.5574	0.7583
0.058	14.0	1694	1.6435	0.7312
0.058	15.0	1815	1.6005	0.7646
0.058	16.0	1936	1.5603	0.7708
0.019	17.0	2057	1.6479	0.7479
0.019	18.0	2178	1.5855	0.7646
0.019	19.0	2299	1.6249	0.7583
0.019	20.0	2420	1.6299	0.7625

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.1+cu121
• Datasets 3.0.1
• Tokenizers 0.19.1